The long?term goal of the Marsh?Armstrong laboratory is to discover biological processes that may someday become treatment targets in Glaucoma. Recently, we discovered that a large fraction of mitochondria are not degraded by the retinal ganglion cells themselves, but rather passed onto astrocytes for degradation. This unusual mode of transcellular degradation of mitochondria, or transmitophagy, occurs more at the optic nerve head than anywhere else in the nervous system. Since the optic nerve head is the likely site of injury in Glaucoma, and a possible site of injury in other optic atrophies that re due to specific deficits in mitochondria, we propose that the degradation of mitochondria at the optic nerve head may be the Achilles' heel of retinal ganglion cells, making them vulnerable to various diseases that manifest as sectorial loss of axons. In the current grant, we propose to test several predictions of this hypothesis. First, we propose to determine whether in rodent models of induced severe mitochondrial impairment, dominant optic atropy and glaucoma, the axonal mitochondria are more susceptible than the mitochondria that are in the soma. Second, we propose to delineate the molecular machinery at the optic nerve head that is responsible for the degradation of axonal mitochondria, and to show that it is indeed critical for the maintenance of healthy retinal ganglion cells. Thirdly, we will determine whether a glaucoma gene, Optineurin, which has been recently shown by others to be involved in mitochondria degradation, acts specifically at the optic nerve head to facilitate the degradation of damaged axonal mitochondria. To test what is admittedly a bold hypothesis, we have developed several new tools that will be applicable to other studies relating to the basic biology of mitochondria, and of how mitochondria are affected in varied diseases. Should the experiments demonstrate that the proposed hypothesis is largely correct, it would represent a significant departure from current views of how glaucomatous axonal loss occurs, and this would have important implications as to how glaucoma might one day be diagnosed and treated. Even if the hypothesis is proven incorrect by the experiments, it will still yield invaluable insight into the biology and pathology of retinal ganglion cells.